The phonology of fortis/lenis in Zapotec in - Natalie Operstein.
The phonology of fortis/lenis in Zapotec in - Natalie Operstein.
The phonology of fortis/lenis in Zapotec in - Natalie Operstein.
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
THE PHONOLOGY OF FORTIS/LENIS IN ZAPOTEC<br />
IN THE LIGHT OF LOANWORDS FROM SPANISH<br />
<strong>Natalie</strong> Operste<strong>in</strong><br />
1. <strong>Zapotec</strong> is a family <strong>of</strong> five to ten (Kaufman n.d.) languages <strong>of</strong> the<br />
Otomanguean stock spoken ma<strong>in</strong>ly <strong>in</strong> the State <strong>of</strong> Oaxaca, Mexico. <strong>The</strong> phonemic<br />
systems <strong>of</strong> most <strong>Zapotec</strong> dialects described to date are based on an opposition between<br />
<strong>fortis</strong> and <strong>lenis</strong> consonants which runs through all or most <strong>of</strong> their obstruents and<br />
sonorants. Even the consonants which are neutral with respect to the <strong>fortis</strong>/<strong>lenis</strong><br />
opposition (which is always the case with the glides and the fricatives /f/, /x ~ h/<br />
borrowed from Spanish, and <strong>of</strong>ten with /r/, one or more affricates, and /m/, which is rare<br />
<strong>in</strong> <strong>Zapotec</strong>) are sometimes said to pattern with either the <strong>fortis</strong> or the <strong>lenis</strong> series (e.g.<br />
Munro and Lopez 1999:2; Regnier 1993:44; Nellis and Hollenbach 1980:92). <strong>The</strong> paired<br />
and unpaired <strong>fortis</strong> and <strong>lenis</strong> consonants <strong>of</strong> a selected number <strong>of</strong> <strong>Zapotec</strong> dialects are<br />
presented <strong>in</strong> Table 1. 1<br />
TABLE 1<br />
Fortis and Lenis Consonants <strong>in</strong> Selected <strong>Zapotec</strong> Dialects<br />
______________________________________________________________________________________<br />
Yateé Zoo Ca YB Mitla Gue SLQZ Isthmus Quieg. Coatlán Texm.<br />
______________________________________________________________________________________<br />
/p/ f f f f f f f f f f f<br />
/b/ l l l l l l l l l l l<br />
/t/ f f f f f f f f f f f<br />
/d/ l l l l l l l l l l l<br />
/k/ f f f f f f f f f f f<br />
/g/ l l l l l l l l l l l<br />
/k/ - - - f f - - - f k f<br />
/g/ - - - l l - - - l l l<br />
/k/ - - - - - - - - f - f<br />
/g/ - - - - - - - - l - l<br />
/ts/ - - f - f f f - - f -<br />
/dz/ - - l - l l - - - - -<br />
/t/ f f f f f f f f f f f<br />
1
d/ l l l l l l - l l l l<br />
/t/ - - - - - f - - - - -<br />
/d/ - - - - - - - - - - -<br />
/s/ f f f f f f f f f f f<br />
/z/ l l l l l l l l l l l<br />
// - f - f f f f f - f f<br />
// - l - l l l l l - l l<br />
// f f f f - f f - f - -<br />
// l l l l - l l - l - -<br />
/f/ - N/A f /loan loan N/A - f loan loan - N/A<br />
/x/ - N/A f /loan loan N/A - f - loan - N/A<br />
/h/ - - - - - - - N/A - - -<br />
// N/A - - - - - - - - - -<br />
/m:/ loan - f - f f f f - - -<br />
/m/ loan N/A - N/A l l l - l? N/A N/A<br />
/n:/ f f f f f f f f - - f<br />
/n/ l l l l l l l l l? N/A l<br />
/:/ - - - - - - - f - - -<br />
// - - - - - - - l - - -<br />
/l:/ f f f f f f f f - - f<br />
/l/ l l l l l l l l l? N/A l<br />
/r:/ - - f - f - f?/loan loan - - -<br />
/r/ loan N/A l N/A l l l? l r? loan N/A<br />
/j/ N/A N/A l N/A N/A N/A l N/A N/A N/A N/A<br />
/w/ N/A N/A l N/A N/A N/A l N/A N/A N/A N/A<br />
‘f’ and ‘l’ are <strong>fortis</strong> and <strong>lenis</strong>, respectively, ‘N/A’ means that the phoneme is not said to participate <strong>in</strong> the<br />
<strong>fortis</strong>/<strong>lenis</strong> dichotomy, ‘-’ that the phoneme is not attested <strong>in</strong> this dialect, ‘loan’ that the phoneme is attested<br />
only <strong>in</strong> Spanish loanwords and has not been considered <strong>in</strong> the light <strong>of</strong> the <strong>fortis</strong>/<strong>lenis</strong> division. ‘Zoo’ is<br />
Zoogocho, ‘Ca’ is Cajonos, ‘YB’ is Yatzachi El Bajo, ‘SLQZ’ is San Lucas Quiav<strong>in</strong>í <strong>Zapotec</strong>, ‘Gue’ is<br />
Guelavía, ‘Quieg.’ is Quiegolani, ‘Texm.’ is San Lorenzo Texmelucan. For Isthmus <strong>Zapotec</strong>, I follow the<br />
phonemic analysis suggested <strong>in</strong> Marlett and Pickett (1987); <strong>in</strong> the <strong>in</strong>ventory given <strong>in</strong> Pickett et al.<br />
(1998:121), /m/ is considered a <strong>lenis</strong> sonorant, there is a <strong>fortis</strong>/<strong>lenis</strong> pair r/rr, and // and /:/ are not cited<br />
as part <strong>of</strong> the phonemic system. Quioquitani <strong>Zapotec</strong>, <strong>in</strong> addition to the obstruents <strong>in</strong> this table, is also said<br />
to have the palatalized <strong>fortis</strong>/<strong>lenis</strong> pairs t/d, c/z, and s/z.<br />
Descriptively, the dist<strong>in</strong>ction between the two series <strong>of</strong> consonants is expressed<br />
differently <strong>in</strong> the sonorants than <strong>in</strong> the different groups <strong>of</strong> obstruents, and depends on the<br />
syllable position <strong>of</strong> the consonant. Correlates <strong>of</strong> <strong>fortis</strong>/<strong>lenis</strong> <strong>in</strong> different dialects spoken<br />
widely apart are <strong>of</strong> a recurr<strong>in</strong>g k<strong>in</strong>d, and can be summarized <strong>in</strong> the form <strong>of</strong> a table: Table<br />
2 shows at a glance differences <strong>in</strong> the behavior <strong>of</strong> <strong>fortis</strong> and <strong>lenis</strong> consonants <strong>in</strong> different<br />
positions. Thus, <strong>fortis</strong> stops and affricates are always voiceless, aspirated word-f<strong>in</strong>ally,<br />
and are never lenited; their <strong>lenis</strong> counterparts show subphonemic variation <strong>in</strong> voic<strong>in</strong>g, no<br />
2
aspiration, and a tendency towards fricativization. Fortis affricates and fricatives are also<br />
said to have greater friction than their <strong>lenis</strong> counterparts. All <strong>fortis</strong> obstruents are said to<br />
be articulated more tensely than the <strong>lenis</strong> ones, and are lengthened after a stressed vowel.<br />
In sonorant consonants, the ma<strong>in</strong> dist<strong>in</strong>guish<strong>in</strong>g feature is the length: <strong>fortis</strong> sonorants are<br />
longer than the <strong>lenis</strong> ones. Fortis lateral is always voiced, while the <strong>lenis</strong> lateral can be<br />
devoiced and accompanied by friction. Both alveolar nasals are voiced, but only the <strong>lenis</strong><br />
one assimilates to the po<strong>in</strong>t <strong>of</strong> articulation <strong>of</strong> the follow<strong>in</strong>g consonant. Word-f<strong>in</strong>ally, the<br />
<strong>lenis</strong> alveolar nasal can be realized as velar or as the nasalization <strong>of</strong> the preced<strong>in</strong>g vowel.<br />
<strong>The</strong> <strong>fortis</strong> alveolar vibrant is a trill, and the <strong>lenis</strong> one is a tap. Until Avel<strong>in</strong>o (2001), all<br />
<strong>lenis</strong> consonants were claimed to lengthen the preced<strong>in</strong>g vowel; Avel<strong>in</strong>o has shown that,<br />
at least <strong>in</strong> the dialect he <strong>in</strong>vestigated, the lengthen<strong>in</strong>g effect is conf<strong>in</strong>ed to the obstruents.<br />
TABLE 2<br />
Correlates <strong>of</strong> Fortis/Lenis <strong>in</strong> Modern <strong>Zapotec</strong> Dialects<br />
______________________________________________________________________________________<br />
Word- Intervocalically Word-<br />
<strong>in</strong>itially after stressed V f<strong>in</strong>ally<br />
______________________________________________________________________________________<br />
f stops/affricates - voice - voice - voice<br />
+ closure + closure + closure<br />
- aspiration - aspiration + aspiration<br />
- lengthen<strong>in</strong>g + lengthen<strong>in</strong>g - lengthen<strong>in</strong>g<br />
l stops/affricates +/- voice +/- voice +/- voice<br />
+/- closure +/- closure +/- closure<br />
- aspiration - aspiration - aspiration<br />
- length - length - length<br />
lengthens preced<strong>in</strong>g stressed V<br />
f fricatives - voice - voice - voice<br />
- lengthen<strong>in</strong>g + lengthen<strong>in</strong>g - lengthen<strong>in</strong>g<br />
l fricatives +/- voice +/- voice +/- voice<br />
- length - length - length<br />
lengthens preced<strong>in</strong>g stressed V<br />
f sonorants + length + length + length<br />
______________________________________________________________________________________<br />
3
<strong>The</strong> correlates <strong>of</strong> <strong>fortis</strong> and <strong>lenis</strong> summarized above and <strong>in</strong> Table 2 received first<br />
experimental confirmation <strong>in</strong> Jaeger’s (1983) study <strong>of</strong> the acoustic properties <strong>of</strong> Yateé<br />
<strong>Zapotec</strong> consonants. This study suggests that the most important feature <strong>of</strong> the <strong>fortis</strong>/<strong>lenis</strong><br />
contrast is likely to be acoustic duration (Jaeger 1983:187-88). A phonetic study <strong>of</strong> the<br />
<strong>fortis</strong>/<strong>lenis</strong> contrast <strong>in</strong> Yalálag <strong>Zapotec</strong> by Avel<strong>in</strong>o (2001) confirms and further elaborates<br />
this conclusion; <strong>in</strong> addition, Avel<strong>in</strong>o discards <strong>in</strong>tensity and articulatory strength as<br />
responsible for the <strong>fortis</strong>/<strong>lenis</strong> contrast, a suggestion repeatedly made <strong>in</strong> the literature on<br />
various <strong>Zapotec</strong> dialects (2001:84-87).<br />
Swadesh (1947) related the <strong>fortis</strong>/<strong>lenis</strong> dichotomy <strong>in</strong> modern <strong>Zapotec</strong> to a<br />
s<strong>in</strong>gle/gem<strong>in</strong>ate dist<strong>in</strong>ction at the Proto-<strong>Zapotec</strong> level. In Swadesh’s view, supported and<br />
further developed <strong>in</strong> Suárez (1973), Kaufman (1983, 1994), and Benton (1988), <strong>lenis</strong><br />
obstruents and the sonorants /l/ and /n/ go back to s<strong>in</strong>gle consonants <strong>in</strong> Proto-<strong>Zapotec</strong>,<br />
while their <strong>fortis</strong> counterparts orig<strong>in</strong>ated <strong>in</strong> gem<strong>in</strong>ate consonants, some <strong>of</strong> which could<br />
have sprung from consonant clusters. As regards the present-day <strong>fortis</strong>/<strong>lenis</strong> pairs m:/m<br />
and r:/r, as well as the Isthmus <strong>Zapotec</strong> pair l:/l, only the <strong>lenis</strong> members <strong>of</strong> each pair go<br />
back to Proto-<strong>Zapotec</strong> sources, while their <strong>fortis</strong> counterparts represent later<br />
developments from other sources. Late creation <strong>of</strong> <strong>fortis</strong> counterparts to these sonorants<br />
is likely to be the result <strong>of</strong> paradigmatic pressure from other <strong>fortis</strong>/<strong>lenis</strong> pairs <strong>in</strong> the<br />
system. <strong>The</strong> immediate source <strong>of</strong> the newly developed <strong>fortis</strong> sonorants seems to have<br />
been compensatory lengthen<strong>in</strong>g, cf. the correlation between Isthmus <strong>Zapotec</strong> <strong>fortis</strong> /l:/<br />
and the length <strong>of</strong> the follow<strong>in</strong>g vowel noted by Benton (1988:17), and that between <strong>fortis</strong><br />
/m:/ <strong>in</strong> SLQZ loanwords and the consonant clusters <strong>in</strong> their Spanish orig<strong>in</strong>als (e.g.<br />
4
zh:ommreel < sombrero ‘hat’, cha’mm < chamba ‘work’, tye’eemm < tiempo ‘time’, and<br />
xtro’oomm < trompo ‘top (toy)’).<br />
<strong>The</strong> tendency to restore the system to symmetry is especially apparent <strong>in</strong> those<br />
dialects <strong>in</strong> which the relationship between the <strong>fortis</strong> and the <strong>lenis</strong> members <strong>of</strong> the<br />
opposition has ceased to be that <strong>of</strong> length. <strong>The</strong> gap between the <strong>fortis</strong> and <strong>lenis</strong> sonorants<br />
is relatively small <strong>in</strong> Zaniza, Texmelucan, and Quioquitani <strong>Zapotec</strong> where, as the result<br />
<strong>of</strong> palatalization <strong>of</strong> Proto-<strong>Zapotec</strong> *nn and *ll, // and // now function as the <strong>fortis</strong><br />
counterparts <strong>of</strong> /n/ and /l/, respectively. In Isthmus <strong>Zapotec</strong>, where the historical result <strong>of</strong><br />
Proto-<strong>Zapotec</strong> *ll is /nd/, the etymological <strong>fortis</strong>/<strong>lenis</strong> pair nd/l is no longer perceived as<br />
such, which has probably contributed to the development <strong>of</strong> the non-etymological <strong>fortis</strong><br />
/l:/ mentioned above (Benton 1988:17). <strong>The</strong> tendency to restore the symmetry <strong>of</strong> the<br />
<strong>fortis</strong>/<strong>lenis</strong> opposition is also apparent <strong>in</strong> the treatment <strong>of</strong> the reflexes <strong>of</strong> Proto-<strong>Zapotec</strong><br />
*ty (Suárez 1973, Kaufman 1983:111, Benton 1988:7-11). In those dialects or<br />
environments where the outcome <strong>of</strong> *ty is a <strong>lenis</strong> stop or affricate, it is symmetrically<br />
matched by the correspond<strong>in</strong>g <strong>fortis</strong> stop or affricate result<strong>in</strong>g from its gem<strong>in</strong>ate<br />
counterpart *tty. In the dialects or environments <strong>in</strong> which this proto-phoneme resulted <strong>in</strong><br />
a rhotic, it is no longer synchronically connectable to its etymological gem<strong>in</strong>ate<br />
counterpart, which has triggered the development <strong>of</strong> a <strong>fortis</strong> alveolar rhotic to fill the<br />
systemic gap (<strong>in</strong> Table 1, such dialects are Mitla, Cajonos, and possibly SLQ <strong>Zapotec</strong>).<br />
Various k<strong>in</strong>ds <strong>of</strong> evidence <strong>in</strong>dicate that <strong>in</strong> the sixteenth century the phonemic<br />
system <strong>of</strong> <strong>Zapotec</strong> already operated on the basis <strong>of</strong> a <strong>fortis</strong>/<strong>lenis</strong> dichotomy. <strong>The</strong> most<br />
important early source <strong>of</strong> evidence is observations <strong>of</strong> Juan de Córdova, the first<br />
missionary grammarian <strong>of</strong> <strong>Zapotec</strong>, on the pronunciation <strong>of</strong> <strong>Zapotec</strong> consonants and the<br />
5
ender<strong>in</strong>g <strong>of</strong> Spanish words by the <strong>Zapotec</strong>s. Another, less direct but no less <strong>in</strong>formative<br />
source is the orthography employed by Córdova <strong>in</strong> his dictionary <strong>of</strong> the same dialect (cf.<br />
Manrique 1966-67; Smith 2000), as well as the orthography <strong>of</strong> other writ<strong>in</strong>gs dat<strong>in</strong>g from<br />
the same period (Broadwell 2000). Yet another source <strong>of</strong> <strong>in</strong>formation on the phonemic<br />
system <strong>of</strong> sixteenth-century <strong>Zapotec</strong> is <strong>Zapotec</strong> render<strong>in</strong>gs <strong>of</strong> numerous Spanish<br />
loanwords that entered its various dialects dur<strong>in</strong>g the course <strong>of</strong> that century.<br />
2. <strong>The</strong> phonemic system <strong>of</strong> sixteenth-century American Spanish has been studied<br />
<strong>in</strong> great detail (for an extensive bibliography on the subject see, e.g., Parodi 1995). For<br />
the phonemic system <strong>of</strong> contemporary <strong>Zapotec</strong> the only source currently available is an<br />
<strong>in</strong>-depth study by Smith (2000) <strong>of</strong> the <strong>phonology</strong> <strong>of</strong> the Valley dialect described by Juan<br />
de Córdova, based ma<strong>in</strong>ly on the orthography employed by Córdova <strong>in</strong> his dictionary <strong>of</strong><br />
this dialect (Córdova 1578b). <strong>The</strong> two phonemic systems are collated below (based on<br />
Parodi 1995:40 and Smith Stark 2000:54).<br />
16 th -c. Spanish 2 16 th -c. Valley <strong>Zapotec</strong><br />
p t t k p t t t k k<br />
b d g b d d d g<br />
z z <br />
f s h s h<br />
m n n <br />
r: mm nn r<br />
λ<br />
l ll<br />
w j w j<br />
6
Some <strong>of</strong> the most salient differences between the above systems <strong>in</strong>clude the fact that<br />
Spanish stops and fricatives constitute voiceless/voiced pairs while the correspond<strong>in</strong>g<br />
<strong>Zapotec</strong> obstruents are divided <strong>in</strong>to <strong>fortis</strong> and <strong>lenis</strong>, and the fact that Spanish has three<br />
series <strong>of</strong> sibilants and <strong>Zapotec</strong> only two. In addition, <strong>Zapotec</strong> has <strong>fortis</strong> and <strong>lenis</strong> versions<br />
<strong>of</strong> each sonorant except the /r/ which, at least accord<strong>in</strong>g to this <strong>in</strong>terpretation <strong>of</strong><br />
Córdova’s <strong>Zapotec</strong>, was absent from the system altogether: <strong>in</strong>stead <strong>of</strong> what appears <strong>in</strong><br />
modern dialects as /r/ Córdova <strong>of</strong>ten writes a , which is <strong>in</strong>terpreted by Smith Stark as<br />
an alveolar stop, or the <strong>lenis</strong> member <strong>of</strong> the pair spelled above as t/d (cf. discussion <strong>of</strong> its<br />
possible surface phonetics <strong>in</strong> Smith 2000: 43-45). 3 F<strong>in</strong>ally, Spanish has a labiodental<br />
fricative which is alien to most <strong>Zapotec</strong> dialects.<br />
3. <strong>The</strong> rema<strong>in</strong>der <strong>of</strong> this paper exam<strong>in</strong>es the treatment <strong>of</strong> Spanish consonants <strong>in</strong><br />
the earliest layer <strong>of</strong> Spanish borrow<strong>in</strong>gs <strong>in</strong> <strong>Zapotec</strong>. <strong>The</strong> earliest layer <strong>of</strong> loanwords is<br />
readily dist<strong>in</strong>guishable from the more recent borrow<strong>in</strong>gs primarily by the treatment <strong>of</strong><br />
Spanish sibilants and the //. For this study, a large number <strong>of</strong> early Spanish loanwordss<br />
<strong>in</strong> various dialects <strong>of</strong> <strong>Zapotec</strong> has been assembled (the dialects exam<strong>in</strong>ed and the<br />
borrowed vocabulary are listed <strong>in</strong> the Appendix). Where there is sufficient data, a<br />
dist<strong>in</strong>ction is made between the three consonantal positions important from the viewpo<strong>in</strong>t<br />
<strong>of</strong> <strong>Zapotec</strong> <strong>phonology</strong>: word-<strong>in</strong>itial, <strong>in</strong>tervocalic, and word-f<strong>in</strong>al. <strong>Zapotec</strong> data are quoted<br />
<strong>in</strong> the orthography <strong>of</strong> the orig<strong>in</strong>al publications.<br />
Stops and /f/<br />
1(a) Spanish p-, b- (spelt , ) 4 > <strong>Zapotec</strong> b-:<br />
7
paño ‘cloth, sash, kerchief’ > Z bay(-ij), T bay, Zoo bay, Ca béy, A payu, YB bey,<br />
Co pai, I bayu’<br />
papaya > MZ baii, YB pey (<strong>in</strong> other dialects borrowed late)<br />
Pedro (name) > Z bed, Ca bε⊥d<br />
peso (a co<strong>in</strong>) > Z bèzh, T peζ&, MZ beex, AZ beψ)u, I beζ&u<br />
barato ‘cheap’ > Z bràd (<strong>in</strong> other dialects borrowed late)<br />
vaca ‘cow’ > Z bàg, MZ baag, SLQZ baag<br />
vigilia ‘vigil’ > Z bixily.<br />
Occasionally, the developments b- > m- or m- > b- are also attested:<br />
batea ‘tray’ > Z (yag-)mtey<br />
botón ‘button’ > Z mu(n)tuny<br />
muñeca ‘doll’ > I buñega’.<br />
1(b) Spanish -p-, -b- (spelt , ) > <strong>Zapotec</strong> -b-:<br />
Felipe (name) > Z lib, SLQZ Li’eb<br />
zapato ‘shoe’ > Z txubat<br />
compadre ‘godfather’ > SLQZ mbaaly, MZ mbaal, I mbale, A umpálí<br />
caballo ‘horse’ > Z kwey, T kΩáy, MZ cabaii, Zoo cabayw, YB cabey, Co wai,<br />
Q gay, SLQZ caba’i<br />
chivo ‘goat’ > Z txib, YB/Zoo σ&ib, SLQZ zhi’eb<br />
navaja ‘fold<strong>in</strong>g knife’ > Z nibàzh, SLQZ nabaazh, MZ nabaax<br />
novillo ‘young bull’ > Z nibily, SLQZ (gùu’ann) nabii.<br />
8
1(c) Spanish f > <strong>Zapotec</strong> p/b (if the names below were <strong>in</strong>deed borrowed early; cf. a<br />
different treatment <strong>of</strong> /f/ <strong>in</strong> the late loan Fransye’scw < Francisco <strong>in</strong> the same dialect):<br />
Florent<strong>in</strong>o > SLQZ Ploory<br />
Alfonsa > SLQZ Po’onnzy<br />
Felix > SLQZ Pu’isy<br />
Epifania > SLQZ Ba’nny.<br />
2(a) Spanish t-, d- > <strong>Zapotec</strong> t-, d-:<br />
taza ‘cup’ > Z tàz, T taz, A taza<br />
teja ‘ro<strong>of</strong><strong>in</strong>g tile’ > I (yoo) deζ&a (yoo ‘house’), SLQZ deezh<br />
tijeras ‘scissors’ > Z tixer, MZ tixer, A tiyera, Q cer (< tser, cf. tmaz < Tomás)<br />
timón ‘beam’ > Z (yag-)tim, SLQZ dye’mm<br />
tomín (a co<strong>in</strong>) > A tummi<br />
testigo > Z testiw, YB testigw, SLQZ testi’u<br />
d<strong>in</strong>ero ‘money’ > T tíny<br />
dom<strong>in</strong>go ‘Sunday’ > Z timiw, MZ dum<strong>in</strong>ngw, Zoo/YB dmigw, SLQZ<br />
Domye’eenngw<br />
durazno ‘peach’ > YB tlas, A trasu, M duras, SLQZ dura’azn.<br />
2(b) Spanish -t-, -d- > <strong>Zapotec</strong> -t-, -d-:<br />
aceite ‘oil’ > Z ased<br />
Antonio (name) > Z Duny, SLQZ Nduuny (cf. also later To’nny)<br />
barato ‘cheap’ > Z bràd<br />
9
limeta ‘bottle’ > Z almet, Zoo lmet, YB lmet, Q lmet<br />
capitán ‘capta<strong>in</strong>’ > Z kaptá<br />
chocolate > Zoo s(i)cwlat, YB σ&cwlat, A choculati, Z txulad, MZ chiculajd,<br />
I dxuladi<br />
morado > Z m(b)ràd (YB morad, Zoo moradw, A moradu are late)<br />
testigo > Z testiw, YB testigw, SLQZ testi’u.<br />
2(c) <strong>The</strong> treatment <strong>of</strong> dj and d / __ i presents a special case: if borrowed early enough,<br />
they fall together with the reflexes <strong>of</strong> Proto-<strong>Zapotec</strong> *ty, cf.:<br />
Dios ‘God’ > Z dyuzh, T nygyooz, SLQZ Dyooz (Zoo Dios and similar forms are<br />
late borrow<strong>in</strong>gs)<br />
medio (a co<strong>in</strong>) > Zoo mechw, YB mech, Ca m:ej<br />
media ‘sock’ > T megy<br />
remedio > YB rmech, SLQZ (Nnambied Dela)rmuudy<br />
sandía > Z x<strong>in</strong>dyi.<br />
In at least one case, Spanish /r/ has a similar treatment:<br />
naranja ‘orange’ > nchaxhu (Diccionario 1995:33).<br />
Cf. the above treatments with the reflexes <strong>of</strong> Proto-<strong>Zapotec</strong> *ty <strong>in</strong> the same languages:<br />
*ke:tyu ‘hole’ (Kaufman 1994:21) > Z gedy, SLQZ guèèe’dy, Zoo yechw,<br />
YB yech<br />
*latyi (tawo) ‘heart’ (Kaufman 1994:20) > Z lady, T (rat) lagy(ã)<br />
*tyowa ‘mouth’ (Kaufman 1994:43) > Z rú’, T rù’, YB cho’a, SLQZ ru’uh,<br />
10
A rú’a.<br />
3(a) Spanish /k/- (spelt , )> <strong>Zapotec</strong> k- or g-:<br />
canoa ‘trough’ > Z kanu<br />
capitán ‘capta<strong>in</strong>’ > Z kaptá<br />
queso ‘cheese’ > Z kèzh, T kyez<br />
cochi ‘pig’ > Z kutx, YB/Zoo coσ&, Co kuucc, A cuttsi, SLQZ cu’uch, MZ cuch<br />
coles ‘cabbages’ > YB corix, A culiyi, Q kliz, SLQZ curehehizh, MZ curijxh 5<br />
cruz ‘cross’ > Z kruz, T kruuz, Zoo cruz, YB coroz, A curuuts, MZ crujz<br />
cuchillo ‘knife’ > MZ guchiil, Zoo cwsiyw, YB cwsiy, A gutsilu.<br />
3(b) Spanish g-, -/k/-, -g- > <strong>Zapotec</strong> g-, -g-:<br />
garbanzo ‘chickpea’ > Z garbaz, SLQZ garba’aannz<br />
garrote ‘stick, staff’ > SLQZ garrood, A yarróté (g- > y- is regular)<br />
vaca ‘cow’ > Z bàg, 6 MZ baag, SLQZ baag<br />
azúcar ‘sugar’ > Z asug<br />
amigo ‘friend’ > Z (a)miw, Zoo/YB migw, SLQZ amiiegw.<br />
<strong>The</strong>re is at least one attested case <strong>of</strong> g- > /k/-, which may have to do with the g- be<strong>in</strong>g<br />
part <strong>of</strong> a cluster:<br />
granada ‘pomegranate’ > Z karnad, SLQZ ca’rnaad.<br />
3(c) Word-f<strong>in</strong>ally, some dialects drop velar stops before rounded vowels:<br />
amigo ‘friend’ > Z (a)miw (but Zoo/YB migw, SLQZ amiiegw)<br />
11
trigo ‘wheat’ > Z triw, MZ triuu, SLQZ tri’u (but Zoo trigw)<br />
yegua ‘mare’ > Z yew, MZ yeuu (but YB yegw)<br />
nigua ‘maggot’ > Z níw, T niw, Co niu, Q niw, SLQZ niuw (but MZ nigw)<br />
banco ‘bank’, ‘bench’ > Z bãw (but SLQZ ba’aanngw, I bangu’)<br />
testigo > Z testiw, SLQZ testi’u (but YB testigw)<br />
dom<strong>in</strong>go > Z timiw (but MZ dum<strong>in</strong>ngw, YB/Zoo dmigw, A dom<strong>in</strong>gu, SLQZ<br />
Domye’eenngw).<br />
(In a later layer <strong>of</strong> loans <strong>in</strong> the same dialects, the velars are preserved <strong>in</strong> this position, cf.<br />
surco ‘furrow’ > Z xurk, T surk, MZ xurc, SLQZ zhu’arc.)<br />
To summarize: (1) voiceless and voiced labial stops are borrowed as <strong>lenis</strong>, both<br />
word-<strong>in</strong>itially and <strong>in</strong>tervocalically. One has to bear <strong>in</strong> m<strong>in</strong>d, however, that word-<strong>in</strong>itial<br />
<strong>fortis</strong> /p/ is extremely rare <strong>in</strong> <strong>Zapotec</strong> (<strong>in</strong> some dialects it is not attested at all, cf. Avel<strong>in</strong>o<br />
2001:6; Nellis and Hollenbach 1980:93), so the <strong>lenis</strong> outcome is expected <strong>in</strong> that<br />
position; (2) Spanish alveolar stops are borrowed as either <strong>fortis</strong> or <strong>lenis</strong> both<br />
<strong>in</strong>tervocalically and word-<strong>in</strong>itially, while /d/ before /i/ is always treated as <strong>lenis</strong>; (3)<br />
Spanish <strong>in</strong>itial /k/- can be borrowed as <strong>fortis</strong> or <strong>lenis</strong>, while <strong>in</strong>tervocalic -/k/-, along with<br />
<strong>in</strong>itial and <strong>in</strong>tervocalic /g/, tends to be borrowed as <strong>lenis</strong>. Thus, only labial Spanish stops<br />
are always borrowed <strong>in</strong>to <strong>Zapotec</strong> as <strong>lenis</strong>. <strong>The</strong> alveolar and velar stops show a great deal<br />
<strong>of</strong> vacillation <strong>in</strong> this respect, but at any rate the pattern <strong>of</strong> their borrow<strong>in</strong>g does not<br />
correlate with their voic<strong>in</strong>g <strong>in</strong> Spanish. While the alveolar stops are borrowed as <strong>fortis</strong> or<br />
<strong>lenis</strong> <strong>in</strong> approximately equal proportion, most <strong>in</strong>itial /k/’s are borrowed as <strong>fortis</strong> and most<br />
<strong>in</strong>tervocalic /k/’s as <strong>lenis</strong>. This perhaps has to do with the frequency <strong>of</strong> distribution <strong>of</strong> the<br />
<strong>fortis</strong> and <strong>lenis</strong> phonemes <strong>in</strong> native <strong>Zapotec</strong> words, or with different correlates <strong>of</strong> the<br />
12
<strong>fortis</strong>/<strong>lenis</strong> dist<strong>in</strong>ction <strong>in</strong> the three series <strong>of</strong> early <strong>Zapotec</strong> stops. Thus, a careful study <strong>of</strong><br />
the treatment <strong>of</strong> stops <strong>in</strong> early Spanish loans <strong>in</strong> <strong>Zapotec</strong> only partially confirms<br />
Kaufman’s (n.d.:18) observation that they, along with other obstruents, were borrowed as<br />
<strong>lenis</strong>.<br />
<strong>The</strong> affricate<br />
In most dialects Spanish /t/ was borrowed as <strong>fortis</strong>, both word-<strong>in</strong>itially and<br />
<strong>in</strong>tervocalically:<br />
chocolate > Z txulad, Zoo s(i)cwlat, YB scwlat, Ca cíkwlát, MZ chiculajd<br />
(but I dxuladi)<br />
chivo ‘goat’ > Z txib, T ciib, Zoo/YB sib, Ca cib (but SLQZ zhi’eb)<br />
coche ‘pig’ > Z kutx, MZ cuch, Zoo/YB cos, Co kuucc, A cuttsi, SLQZ cu’uch<br />
cuchara ‘spoon’ > Zoo cwsar, YB c(w/o)sar (but SLQZ wzhyaar, I (g)udxara)<br />
cuchillo ‘knife’ > MZ guchiil, Zoo cwsiyw, YB cwsiy, Z gutsilu, SLQZ bchiilly<br />
(but I (g)udxíu)<br />
macho ‘mule’ > Z matx, T mac, Zoo masw, YB mas (but MZ madz)<br />
machete > Zoo maset, YB mset, Co macctt (but MZ madxed, SLQZ<br />
mazhye’edy, Q mzæd)<br />
mecha ‘wick’ > T mec, Ca m:ec. 7<br />
Reflexes <strong>of</strong> this affricate <strong>in</strong> early loanwords co<strong>in</strong>cide with those <strong>of</strong> Proto-<strong>Zapotec</strong><br />
gem<strong>in</strong>ate *cc <strong>in</strong> native morphemes, cf.:<br />
13
*ccho-n(n)a ‘three’ (Benton 1988 #122 ) > Z txun, T con, Zoo sone, YB son, A<br />
tsunná, Ca cón:é, MZ chon<br />
Sibilants<br />
*kiccha ‘hair’ (Benton 1988 #55)/*kittza(7) (Kaufman 1994:20) > Z gitx, T gyìc,<br />
Zoo yisa’, YB yis’, A íttsa’ (íqquia), SLQZ gyihch, Co kìcc.<br />
1(a) Spanish /s/ (spelt , ) and /z/ (spelt ) 8 > <strong>Zapotec</strong> /z/ and /s/:<br />
arroz ‘rice’ > Z arùz, YB roz (Zoo ros, SLQZ rro’s are late)<br />
azúcar ‘sugar’ > Z asug (MZ su’cr, Zoo/YB sucr, SLQZ sua’rc may be late)<br />
ciudad ‘city’ > Z siwda, Zoo ciuda, YB syoda, SLWZ syudaa<br />
coc<strong>in</strong>ero ‘cook’ > Z kusnely<br />
cruz ‘cross’ > Z cruz, Zoo cruz, YB coroz, MZ crujz (SLQZ cru’uhsy may be<br />
late)<br />
durazno ‘peach’ > YB tlas, A trasu, SLQZ dura’azn, MZ duras (all may be late)<br />
garbanzo ‘chickpea’ > Z garbaz, SLQZ garba’aannz, MZ garbans (late)<br />
mazo ‘mallet’ > Z mez, T mãz, Zoo/YB maz, SLQZ maaz, MZ mas (late)<br />
mostaza ‘mustard’ > MZ (yag) muxtas (yag ‘tree’)<br />
mozo ‘servant’ > Z mùz (MZ/Zoo/YB mos is late).<br />
It is likely that part or all <strong>of</strong> the loanwords <strong>in</strong> which Spanish /s/ <strong>of</strong> /z/ > <strong>Zapotec</strong> /s/ were<br />
borrowed later than those <strong>in</strong> which Spanish /s/ or /z/ > <strong>Zapotec</strong> /z/.<br />
1(b) <strong>The</strong> treatment <strong>of</strong> <strong>in</strong> two loanwords can be taken as an <strong>in</strong>dication <strong>of</strong> its affricated<br />
pronunciation <strong>in</strong> Spanish. Spanish zapato ‘shoe’ was borrowed <strong>in</strong> Zaniza <strong>Zapotec</strong> as<br />
txubat, and Spanish cruz ‘cross’ was borrowed <strong>in</strong> Atepec <strong>Zapotec</strong> as curuuts. Zaniza tx<br />
14
and Atepec ts normally render Spanish ch (cf. chivo ‘goat’ > Z txib and cuchillo ‘knife’ ><br />
A gutsilu). <strong>The</strong> treatment <strong>of</strong> <strong>in</strong> these loans is therefore consistent with its affricated<br />
pronunciation (/ts/) <strong>in</strong> Spanish, and may be taken as evidence for a late survival <strong>of</strong> this<br />
affricate.<br />
2. Spanish // (spelt , ) and /⏐ / (spelt ) > <strong>Zapotec</strong> // or //:<br />
sacristán > Z xundista, A yueda(yoto’) [yoto’ ‘temple, church’], SLQZ<br />
sacax:taany<br />
san, santo ‘sa<strong>in</strong>t’ > I (beu) zandu’; za(bizende) ‘San Vicente’, SLQZ Xann(daan)<br />
‘Santa Ana’; Xmo’oony ‘Santa Mónica’<br />
semana > Z ximan, YB sman, Zoo xman, A yumanu, Ca zm:an, SLQZ/MZ<br />
xmaan<br />
silla ‘chair’, ‘saddle’ > Z xily, T sily, MZ (yag)xhil, SLQZ zhi’iilly, A xila’<br />
sombrero > MZ xhumbreel, SLQZ zh:ommreel<br />
escuela > Z xikwal<br />
camisa > Z mìzh, MZ (re-)gamizh, A miya, I gamiza<br />
manso > Z màzh, T maz, YB max, MZ madx<br />
misa ‘mass’ > Z mìzh, T míz, I míza’ , MZ mix, A miya (YB/Zoo mis, SLQZ<br />
mye’es are late)<br />
peso > Z bezh, T pez, MZ beex, A beyu, I bezu<br />
Dios > Z dyuzh, T nygyooz, A (Tata) Diuy(a), I dyuzi, Q dyuz (< adiós)<br />
Tomás > Z màzh, SLQZ Ma’azhy, Q tmaz<br />
15
Luis > Z wizh, T wiiz.<br />
In the dialects that currently dist<strong>in</strong>guish (and probably did so <strong>in</strong> the sixteenth century)<br />
between alveopalatal and retr<strong>of</strong>lex sibilants, Spanish // and // were borrowed, with<br />
very few exceptions, as alveopalatal.<br />
3. Spanish // (spelt ) and // (spelt and ) > <strong>Zapotec</strong> // or //:<br />
jabón ‘soap’ > I zabú<br />
jarro ‘pitcher’ > A yaru(iyya) [iyya ‘flower’]<br />
jer<strong>in</strong>ga ‘syr<strong>in</strong>ge’ > I zir<strong>in</strong>ga<br />
jícara ‘calabash cup’ > Z xìg, I ziga, SLQZ zh:i’ahg, MZ xijg, YB xigu’<br />
Juana (name) > Ca zwán, SLQZ Zh:ùaan<br />
gigante > Z xigan<br />
aguja ‘needle’ > A gúψ)á, MZ guux, SLQZ (guìi’ch)gwu’ùa’zh:<br />
ajo ‘garlic’ > Z àzh, T az, Zoo/YB (cuan)ax, A gayu, SLQZ (xti)aazh, MZ aax<br />
arveja ‘pea’ > A (daa)ribeyi (daa ‘beans’)<br />
clavija ‘peg’ > Z (yag-)kabizh (yag ‘wood’), SLQZ garbiizh<br />
mixe ‘Mixe’ > MZ miix, YB/Zoo mix, SLQZ Miìi’zh<br />
naranja > T láz, A maraya, SLQZ nraazh, MZ naraax<br />
navaja > Z nibazh, SLQZ nabaazh, MZ nabaax<br />
tijeras ‘scissors’ > Z tixer, MZ tixer, Q cer (< tser), A tiyera, SLQZ (gyìe’b)<br />
zhiier.<br />
16
In dialects that currently dist<strong>in</strong>guish between alveopalatal and retr<strong>of</strong>lex sibilants, //<br />
and // were borrowed as alveopalatal.<br />
To summarize: (1) the affricate ch is mostly borrowed as <strong>fortis</strong>; (2) /s/ (and /z/)<br />
are mostly borrowed as <strong>lenis</strong>, but <strong>in</strong> Zaniza zapato and Atepec cruz they are rendered like<br />
Spanish ch and may reflect an affricated pronunciation <strong>of</strong> Spanish ; (3) Spanish<br />
//, //, // and // are borrowed as <strong>fortis</strong> or <strong>lenis</strong> alveopalatal fricatives (// and //) word-<br />
<strong>in</strong>itially, and, with one exception, as <strong>lenis</strong> (//) <strong>in</strong>tervocalically. Thus, the pattern <strong>of</strong><br />
render<strong>in</strong>g obstruents <strong>in</strong> Spanish loans that emerges is a complex one. As can be seen<br />
from Table 3, the voic<strong>in</strong>g <strong>of</strong> Spanish obstruents does not affect the borrow<strong>in</strong>g pattern,<br />
which <strong>in</strong>stead seems to depend on other factors.<br />
TABLE 3<br />
Render<strong>in</strong>g <strong>of</strong> Spanish Obstruents <strong>in</strong> Early <strong>Zapotec</strong><br />
______________________________________________________________________________________<br />
Word-<strong>in</strong>itially Word-medially Word-f<strong>in</strong>ally<br />
______________________________________________________________________________________<br />
p, b (, ) l l not attested<br />
t, d f/l f/l not attested<br />
d+i l l not attested<br />
/k/ f/l l not attested<br />
g l l not attested<br />
ch f f not attested<br />
/s/, /z/ l l l<br />
//, //, //, // f/l l l<br />
4. <strong>The</strong> <strong>phonology</strong> <strong>of</strong> Córdova’s <strong>Zapotec</strong> based on a study <strong>of</strong> his orthography<br />
(Smith Stark 2000) can be <strong>of</strong> some assistance <strong>in</strong> elucidat<strong>in</strong>g the above borrow<strong>in</strong>g pattern.<br />
<strong>The</strong> follow<strong>in</strong>g traits <strong>of</strong> Córdova’s orthography appear to be relevant: (1) there seem to be<br />
no examples <strong>of</strong> word-<strong>in</strong>itial <strong>fortis</strong> /p/; (2) only the letters used for render<strong>in</strong>g voiceless<br />
obstruents <strong>in</strong> Spanish are used <strong>in</strong> render<strong>in</strong>g <strong>fortis</strong> obstruents <strong>in</strong> <strong>Zapotec</strong>; (3) the letters<br />
17
used for voiced obstruents <strong>in</strong> Spanish can be used <strong>in</strong> render<strong>in</strong>g <strong>Zapotec</strong> <strong>lenis</strong> segments<br />
/b/, /g/, /z/, and /Ζ/; (4) both <strong>fortis</strong> and <strong>lenis</strong> dental stops are represented by the letter ;<br />
(5) <strong>fortis</strong> obstruents <strong>in</strong> Córdova’s <strong>Zapotec</strong> were apparently lengthened <strong>in</strong> posttonic<br />
syllables s<strong>in</strong>ce the <strong>in</strong>frequent examples <strong>of</strong> double spell<strong>in</strong>g <strong>of</strong> consonants are conf<strong>in</strong>ed to<br />
this position; (6) <strong>Zapotec</strong> /s/ and /z/ are reasonably well differentiated graphically, but<br />
// and // are not (Smith 2000:32-33, 35-46). Additional <strong>in</strong>formation on the <strong>fortis</strong>/<strong>lenis</strong><br />
<strong>phonology</strong> <strong>of</strong> Córdova’s <strong>Zapotec</strong> may be found <strong>in</strong> his remarks <strong>in</strong> the Arte. Here, Córdova<br />
enumerates such mistakes made, he says, mostly by Spaniards <strong>in</strong> their <strong>Zapotec</strong>, as<br />
pronounc<strong>in</strong>g <strong>in</strong>tervocalic /b/ for /p/, /g/ for /k/ and vice versa, /z/ for /s/, <strong>in</strong>itial /d/ for<br />
Spanish /t/ (‘Doledo’ for ‘Toledo’), and (alveopalatal sibilant) for (retr<strong>of</strong>lex<br />
sibilant) (1578a:73). While it is likely that some <strong>of</strong> these observations reflect more on<br />
Córdova’s Spanish than on his <strong>Zapotec</strong>, they seem to expla<strong>in</strong> (1) the absence <strong>of</strong> the <strong>in</strong>itial<br />
<strong>fortis</strong> /p/ <strong>in</strong> render<strong>in</strong>g Spanish loans (expla<strong>in</strong>able by the absence <strong>of</strong> <strong>in</strong>itial /p/ <strong>in</strong> native<br />
words), (2) the little role played by the voic<strong>in</strong>g <strong>of</strong> Spanish obstruents <strong>in</strong> the overall<br />
borrow<strong>in</strong>g pattern and especially <strong>in</strong> that <strong>of</strong> the dental stops, and (3) the absence <strong>of</strong><br />
render<strong>in</strong>gs <strong>of</strong> <strong>in</strong>tervocalic Spanish /p/ and /k/ as <strong>fortis</strong> (expla<strong>in</strong>able by the lengthen<strong>in</strong>g <strong>of</strong><br />
the native <strong>fortis</strong> obstruents <strong>in</strong> this position). In addition, Córdova’s observations on the<br />
pronunciation mistakes that <strong>in</strong>volve voiced and voiceless alveolar sibilants and only the<br />
voiceless alveopalatal sibilant probably po<strong>in</strong>t to the voicelessness <strong>of</strong> // <strong>in</strong> his Spanish,<br />
which <strong>in</strong> turn expla<strong>in</strong>s the under-differentiation <strong>of</strong> // and // <strong>in</strong> his <strong>Zapotec</strong> orthography.<br />
5. Nasals<br />
1(a) Spanish m > m <strong>in</strong> <strong>Zapotec</strong>:<br />
18
macho ‘mule’ > Z matx, T mac, MZ madz, Zoo masw, YB mas<br />
mazo ‘mallet’ > Z mez, Tmãz, Zoo/YB maz, SLQZ maaz, MZ mas<br />
mula ‘female mule’ > Z/T muly, MZ mul, SLQZ muuall<br />
almohada ‘pillow’ > Z almàd, YB lmad, SLQZ almwaad<br />
amigo ‘friend’ > Z (a)miw, YB/Zoo migw, SLQZ amiiegw<br />
limeta ‘bottle’ > Z almet, Zoo lmet, YB lmet<br />
compadre ‘godfather’ > A umpali, SLQZ mbaaly, MZ mbaal, I mbale<br />
comadre ‘godmother’ > I male, Ca m:ál<br />
dom<strong>in</strong>go > Z timiw, MZ dum<strong>in</strong>ngw, Zoo/YB dmigw, Z dom<strong>in</strong>gu, SLQZ<br />
Domye’eenngw<br />
semana ‘week’ > Z ximan, YB σ&man, Zoo xman, A yumanu, SLQZ/MZ xmaan.<br />
1(b) Cases where Spanish /m/ was borrowed as <strong>fortis</strong> have to do with the loss <strong>of</strong> a<br />
follow<strong>in</strong>g consonant and subsequent compensatory lengthen<strong>in</strong>g <strong>of</strong> the nasal:<br />
sombrero ‘hat’ > SLQZ zh:ommreel<br />
timón ‘beam’ > SLQZ dye’mm<br />
tiempo ‘time’ > SLQZ tye’eemm<br />
trompo ‘top’ (toy) > SLQZ xtro’oomm<br />
tomín (co<strong>in</strong>) > A tummi.<br />
1(c) Some /m/’s <strong>in</strong> <strong>Zapotec</strong> loans resulted from other consonants:<br />
batea ‘tray’ > Z (yag-)mtey<br />
botón ‘button’ > Z mu(n)tuny<br />
19
naranja ‘orange’ > A maraya, Ca m:raz, mbrhaxh (Diccionario 1995:33).<br />
2(a) Spanish n- > <strong>lenis</strong> n- <strong>in</strong> <strong>Zapotec</strong>:<br />
naranja ‘orange’ > SLQZ nraazh, MZ naraax<br />
navaja ‘fold<strong>in</strong>g knife’ > Z nibazh, SLQZ nabaazh, MZ nabaax<br />
nigua ‘maggot’ > Z niw, Co niu, SLQZ niuw, MZ nigw, Q niw<br />
novillo ‘young bull’ > Z nibily, SLQZ (gùu’ann) nabii.<br />
2(b) Spanish -n- > <strong>fortis</strong> or <strong>lenis</strong> -n- <strong>in</strong> <strong>Zapotec</strong>:<br />
d<strong>in</strong>ero ‘money’ > T tíny<br />
m<strong>in</strong>a ‘m<strong>in</strong>e’ > T m<strong>in</strong>y<br />
panela ‘sugar loaf’ > Z p<strong>in</strong>yal<br />
semana ‘week’ > Z ximan, YB sman, Zoo xman, A yumanu, SLQZ/MZ xmaan.<br />
Spanish -n- tends to be lost before fricatives, but is preserved before stops:<br />
garbanzo ‘chickpea’ > Z garbaz, SLQZ garba’aannz<br />
manso ‘tame’ > Z màzh, T maz, YB max, MZ madx<br />
naranja ‘orange’ > A maraya, Ca m:raz, T láz, SLQZ nraazh, MZ naraax<br />
sandía ‘watermelon’ > Z x<strong>in</strong>dyi, SLQZ xanndiia, I zandie’<br />
culantro ‘cilantro’ > Z kulyandr, A culandru, SLQZ cura’aann<br />
banco ‘bank’, ‘bench’ > SLQZ ba’aanngw, Z bãw, Zoo/YB bancw, I bangu’<br />
dom<strong>in</strong>go ‘Sunday’ > MZ dum<strong>in</strong>ngw, SLQZ Domye’eenngw, A dom<strong>in</strong>gu.<br />
20
2(c) Word-f<strong>in</strong>al Spanish -n can be rendered by a <strong>fortis</strong> or <strong>lenis</strong> nasal, or be lost,<br />
depend<strong>in</strong>g on the dialect:<br />
sacristán > Z xundista, A yueda(yoto’), SLQZ sacax:taany<br />
timón ‘beam’ > Z (yag-)tim, SLQZ dye’mm (-mm < *-mn; note also the stress<br />
shift)<br />
tomín (a co<strong>in</strong>) > A tummi<br />
botón ‘button’ > Z mu(n)tuny, SLQZ btoony.<br />
3(a) Spanish ñ > <strong>lenis</strong> n <strong>in</strong> <strong>Zapotec</strong>:<br />
albañil ‘mason’ > Z arbanyil, YB/Zoo albanil<br />
escaño ‘bench with a back’ > YB/Zoo xcan, MZ xcaan.<br />
3(b) <strong>The</strong>re is also one example <strong>of</strong> an early loan <strong>in</strong> which Spanish ñ > <strong>Zapotec</strong> y:<br />
paño ‘cloth’ > Z bay(-ij), T bay, I bayu’, Co pai, A payu, YB bey, Zoo bay.<br />
<strong>The</strong> uniform outcome <strong>of</strong> ñ as a glide, despite the fact that Córdova’s and doubtless other<br />
dialects had or could have had a palatal nasal match<strong>in</strong>g the Spanish ñ seems to argue for a<br />
borrow<strong>in</strong>g through the medium <strong>of</strong> Nahuatl, where y is one <strong>of</strong> the possible outcomes <strong>of</strong><br />
early Spanish ñ, and paño is reflected as payo (cf. González Casanova 1977:131, 146).<br />
Liquids<br />
1(a) Spanish l- > <strong>fortis</strong> or <strong>lenis</strong> l- <strong>in</strong> <strong>Zapotec</strong>:<br />
lazo > T laz<br />
limeta ‘bottle’ > Z almet, Zoo lmet, YB lmet, Q lmet<br />
lunes ‘Monday’ > Z lunex, MZ lun, Zoo lun, YB (zha) lon, A luni, SLQZ Luuny<br />
21
Lucas (name) > Z lyuj, T luk, SLQZ Lu’c.<br />
1(b) Spanish -l- > <strong>fortis</strong> or <strong>lenis</strong> -l- <strong>in</strong> <strong>Zapotec</strong>:<br />
alguacilillo (dim. <strong>of</strong> alguacil ‘constable’) > SLQZ lasliiery<br />
culantro ‘cilantro’ > Z kulyandr (YB culantr, A culandru)<br />
chocolate > Z txulad, Zoo s (i)cwlat, YB σ&cwlat, MZ chiculajd<br />
escuela ‘school’ > Z xikwal<br />
mezcal ‘agave liquor’ > Z mixcaly (Zoo mezcal, YB mescal, SLQZ mescaaly)<br />
mole ‘stew with chili sauce’ > SLQZ mo’lly, MZ moll<br />
mula ‘female mule’ > Z/T muly, MZ mul, SLQZ muuall<br />
panela ‘sugar loaf’ > Z p<strong>in</strong>yal (YB/Zoo panel, MZ paneel)<br />
real (a co<strong>in</strong>) > YB ryel ~ riel, A rriali, SLQZ rryeelly, MZ räjl<br />
vigilia ‘vigil’ > Z bixily<br />
Manuel (name) > Z wely, SLQZ Ne’ll<br />
Pablo (name) > T baly<br />
Samuel (name) > Z wely, T mel.<br />
1(c) <strong>Zapotec</strong> <strong>fortis</strong> or <strong>lenis</strong> /l/ can result from other Spanish consonants:<br />
from /r/:<br />
coc<strong>in</strong>ero ‘cook’ > Z kusnely<br />
compadre ‘godfather’ > A umpali, SLQZ mbaaly, MZ mbaal, I mbale<br />
comadre ‘godmother’ > I male<br />
durazno ‘peach’ > YB tlas<br />
22
from /d/:<br />
naranja ‘orange’ > T láz<br />
sombrero ‘hat’ > A umbrelu, MZ xhumbreel, SLQZ zh:ommreel;<br />
medio (a co<strong>in</strong>) > SLQZ mùuully, MZ meel (and probably A belliu).<br />
2(a) Spanish ll [×] > <strong>fortis</strong> or <strong>lenis</strong> l <strong>in</strong> <strong>Zapotec</strong>:<br />
llave ‘key’ > SLQZ lye’i, MZ liäii<br />
cuchillo ‘knife’ > MZ guchiil, A gutsilu, SLQZ bchiilly (but YB cwshiy,<br />
Zoo cwshiyw, I gudxíu)<br />
Castilla ‘Castile’ > ZooZ (dizha’)xtil,YB (dizhe’e)xtil, A (la’a)xtila, MZ<br />
(didx)xtiil ‘Spanish (language)’ 9<br />
manzanilla ‘camomile’ > MZ maNsanil (N = <strong>fortis</strong> /n/)<br />
mol<strong>in</strong>illo ‘hand mill’ > SLQZ mo/urniilly, MZ morniil<br />
novillo ‘young bull’ > Z nibily<br />
silla ‘chair’, ‘saddle’ > Z xily, T σ&ily, MZ (yag)xhil, SLQZ zhi’iilly, A xila’.<br />
2(b) A couple <strong>of</strong> loans provide evidence <strong>of</strong> early yeísmo (i.e. the pronunciation <strong>of</strong><br />
Spanish ll as [y]):<br />
caballo ‘horse’ > Z kwey, at káy, Co wai, YB cabey, MZ cabaii, Zoo cabayw,<br />
SLQZ caba’i, Q gay<br />
pollo ‘chicken’ > Co poi, I buyu’.<br />
3(a) Spanish r- (a trill) > <strong>Zapotec</strong> rr-/r- (one example):<br />
23
eal (a co<strong>in</strong>) > YB ryel ~ riel, A rriali, SLQZ rryeelly, MZ räjl.<br />
3(b) Spanish -rr- and -r- > <strong>Zapotec</strong> (<strong>lenis</strong>) r:<br />
arroz ‘rice’ > Z arùz, YB roz<br />
barato ‘cheap’ > Z bràd (also MZ bará’t, Zoo baratw, YB barat, SLQZ baraa’t)<br />
morado ‘purple’ > Z m(b)ràd (also YB morad, Zoo moradw, A moradu)<br />
naranja ‘orange’ > A maraψ)a, SLQZ nraazh, MZ naraax<br />
Andrés (name) > MZ ndré(h)zh.<br />
3(c) <strong>Zapotec</strong> /r/ can also result from Spanish /l/ and //:<br />
alguacilillo (dim. <strong>of</strong> alguacil ‘constable’) > SLQZ lasliiery<br />
albañil ‘mason’ > Z arbanyil<br />
alcalde ‘mayor’ > Zoo rcal, YB rcal, SLQZ rca’alldy<br />
clavija ‘beam’ > SLQZ garbiizh, Z (yag-)kabizh<br />
coles ‘cabbages’ > YB corix, Ca kórìσ&, SLQZ curehehizh, MZ curijxh<br />
culantro ‘cilantro’ > SLQZ cura’aann<br />
mol<strong>in</strong>illo ‘hand mill’ > SLQZ mo/urniilly, MZ morniil.<br />
In some cases (as <strong>in</strong> albañil, alcalde, mol<strong>in</strong>illo) this development can be expla<strong>in</strong>ed by<br />
dissimilation.<br />
In one case, <strong>Zapotec</strong> -r- has probably resulted from Zpanish -d-:<br />
maravedí (an old co<strong>in</strong>) > Z mrí ‘money’, T mbrii ‘six centavos’. 10<br />
To summarize the situation with the borrow<strong>in</strong>g <strong>of</strong> sonorants: (1) Spanish /m/ is<br />
borrowed as <strong>fortis</strong> only when the loss <strong>of</strong> the follow<strong>in</strong>g consonant causes its compensatory<br />
24
lengthen<strong>in</strong>g; (2) <strong>in</strong>itial /n/ is borrowed as <strong>lenis</strong>, while <strong>in</strong>tervocalic and word-f<strong>in</strong>al /n/ may<br />
be borrowed as <strong>lenis</strong> or <strong>fortis</strong>. This situation may have someth<strong>in</strong>g to do with the<br />
distribution <strong>of</strong> the <strong>fortis</strong> and <strong>lenis</strong> /n/ <strong>in</strong> native <strong>Zapotec</strong> words: thus, Benton (1988:15-16)<br />
does not reconstruct word-<strong>in</strong>itial *nn- or <strong>in</strong>tervocalic *-n- <strong>in</strong> his version <strong>of</strong> Proto-<br />
<strong>Zapotec</strong>; (3) Spanish ñ is borrowed as <strong>lenis</strong> /n/, with the exception <strong>of</strong> the word paño,<br />
likely to have been borrowed through Nahuatl, where Spanish ñ > <strong>Zapotec</strong> y; (4) Spanish<br />
/l/ and // are borrowed as either <strong>fortis</strong> or <strong>lenis</strong> /l/ and occasionally as an /r/; (5) at least<br />
two loanwords provide evidence for early yeísmo; (6) Spanish trilled /r/ is occasionally<br />
borrowed as <strong>fortis</strong> <strong>in</strong> the dialects that dist<strong>in</strong>guish between <strong>fortis</strong> and <strong>lenis</strong> /r/; (7) <strong>in</strong> a<br />
number <strong>of</strong> cases, Spanish /r/ or /d/ have been borrowed as <strong>Zapotec</strong> <strong>fortis</strong> or <strong>lenis</strong> /l/.<br />
6. In the loanwords that only recently entered <strong>Zapotec</strong>, the pattern <strong>of</strong> borrow<strong>in</strong>g<br />
has changed considerably. First, recent loans naturally reflect changes undergone by the<br />
<strong>phonology</strong> <strong>of</strong> American Spanish s<strong>in</strong>ce the sixteenth century. One <strong>of</strong> these is the<br />
velarization <strong>of</strong> the palatal sibilant //, reflected <strong>in</strong> early loans such as A (daa)ribeyi <<br />
arveja ‘pea’, to /x/. Second, prolonged exposure to the phonological system <strong>of</strong> Spanish<br />
due to extensive lexical borrow<strong>in</strong>g and massive bil<strong>in</strong>gualism have exercised a powerful<br />
<strong>in</strong>fluence on the borrow<strong>in</strong>g strategies <strong>of</strong> the <strong>in</strong>dividual dialects and <strong>of</strong> <strong>Zapotec</strong> as a<br />
whole. Changes <strong>in</strong> the borrow<strong>in</strong>g pattern are partly due to the <strong>in</strong>troduction <strong>of</strong> the<br />
phonemes /f/, /x/ (or /h/), and <strong>in</strong> some cases /r/ (e.g., <strong>in</strong> Coatlan <strong>Zapotec</strong>, cf. Rob<strong>in</strong>son<br />
1963) <strong>in</strong> the phonological systems <strong>of</strong> <strong>Zapotec</strong> dialects. Thus, while <strong>in</strong> earlier loans both<br />
/f/ and /x/ were rendered by stops (cf. YB lberg, SLQZ albe’erg < arveja ‘pea’), <strong>in</strong> more<br />
recent borrow<strong>in</strong>gs they are borrowed as fricatives (cf. MZ alberj ‘pea’). 11 Exposure to<br />
25
Spanish has also re<strong>in</strong>forced contrasts that existed <strong>in</strong> <strong>Zapotec</strong> only at a subphonemic level,<br />
trigger<strong>in</strong>g rearrengements <strong>in</strong> the distribution <strong>of</strong> native phonemes. Both types <strong>of</strong> <strong>in</strong>fluence<br />
can be observed when compar<strong>in</strong>g the treatment <strong>of</strong> Spanish consonants <strong>in</strong> the early and the<br />
more recent strata <strong>of</strong> borrowed vocabulary. Thus, <strong>in</strong> Cajonos <strong>Zapotec</strong> both voiced and<br />
voiceless Spanish obstruents are usually reflected <strong>in</strong> early loans as <strong>lenis</strong> (e.g. béy < paño,<br />
bd < Pedro); <strong>in</strong> recent loans Spanish voiceless stops are borrowed as <strong>fortis</strong>, and voiced<br />
stops as <strong>lenis</strong>. This borrow<strong>in</strong>g pattern has caused <strong>fortis</strong> /p/ to appear <strong>in</strong> word-<strong>in</strong>itial<br />
position, which it never does <strong>in</strong> native words, and has also led to a greater importance <strong>of</strong><br />
voic<strong>in</strong>g <strong>in</strong> contrast<strong>in</strong>g the two series <strong>of</strong> obstruents (cf. Nellis and Hollenbach 1980:93,<br />
104-05). 12 /r/ and /l/ <strong>in</strong> the same dialect sometimes replace each other <strong>in</strong> early borrow<strong>in</strong>gs<br />
(e.g. m:ál < comadre ‘godmother’, kórìs < coles ‘cabbages’), but <strong>in</strong> recent loans they are<br />
borrowed as /r/ and /l/, respectively, possibly due the <strong>in</strong>troduction <strong>of</strong> a greater contrast<br />
between the two liquids ow<strong>in</strong>g to a steady flow <strong>of</strong> Spanish loans. Also, while <strong>in</strong> early<br />
loans Spanish trilled /r/ was borrowed as <strong>lenis</strong>, <strong>in</strong> later loans it can be rendered <strong>in</strong> <strong>Zapotec</strong><br />
by a <strong>fortis</strong> rhotic, cf. címa r < chamarra ‘blanket’ (an old loan) versu r:ey < raya ‘l<strong>in</strong>e’ (a<br />
recent loan). It has already been hypothesized above that <strong>fortis</strong> /r/ is a recent <strong>in</strong>novation<br />
<strong>in</strong> some <strong>Zapotec</strong> dialects, conceivably triggered by a synchronic dissociation between<br />
<strong>lenis</strong> /r/ and its etymological <strong>fortis</strong> counterpart. It is also likely that this <strong>Zapotec</strong>-<strong>in</strong>ternal<br />
development received additional re<strong>in</strong>forcement from the existence <strong>of</strong> two rhotics <strong>in</strong><br />
Spanish loanwords.<br />
Dialects other than Cajonos <strong>Zapotec</strong> show comparable adjustments <strong>in</strong> their<br />
borrow<strong>in</strong>g strategies, generally <strong>in</strong> the direction <strong>of</strong> a greater attun<strong>in</strong>g to the <strong>phonology</strong> <strong>of</strong><br />
Spanish. Among these may be mentioned the change <strong>in</strong> the render<strong>in</strong>g <strong>of</strong> Spanish /n/ and<br />
26
l/: while <strong>in</strong> early loans these could be rendered with both <strong>fortis</strong> and <strong>lenis</strong> native<br />
phonemes without a discernible distributional pattern, <strong>in</strong> later loans they are mostly<br />
borrowed as <strong>lenis</strong>. Some differences <strong>in</strong> the treatment <strong>of</strong> Spanish consonants <strong>in</strong> the early<br />
and late loans are summarized <strong>in</strong> Table 4.<br />
TABLE 4<br />
Changes <strong>in</strong> the Pattern <strong>of</strong> Consonant Borrow<strong>in</strong>g<br />
______________________________________________________________________________________<br />
Spanish consonants Early loans Recent loans<br />
______________________________________________________________________________________<br />
p-b, t-d, k-g borrowed as f or l; borrowed as f or l;<br />
voic<strong>in</strong>g unimportant borrow<strong>in</strong>g pattern<br />
for distribution based on voic<strong>in</strong>g<br />
s borrowed as l borrowed as f<br />
f borrowed as /p/, /b/ borrowed as /f/<br />
j [x/h] borrowed /g/ borrowed as [x/h]<br />
m borrowed as /m/ borrowed as /m/<br />
n borrowed as f or l borrowed as l<br />
l borrowed as f or l borrowed as l<br />
r borrowed as l borrowed as l<br />
rr borrowed as l borrowed as f or l<br />
7. This paper has exam<strong>in</strong>ed the adaptation <strong>of</strong> Spanish loanwords <strong>in</strong> various<br />
<strong>Zapotec</strong> dialects. Although the ma<strong>in</strong> focus <strong>of</strong> the paper has been early loans, changes <strong>in</strong><br />
the borrow<strong>in</strong>g strategy that occurred between the earliest and the more recent layers <strong>of</strong><br />
loans have also been considered. A comparison <strong>of</strong> the borrow<strong>in</strong>g patterns dur<strong>in</strong>g these<br />
periods <strong>in</strong>dicates that contact with Spanish has exercized a considerable <strong>in</strong>fluence on<br />
<strong>Zapotec</strong> <strong>phonology</strong>, the most salient elements <strong>of</strong> which are the <strong>in</strong>troduction <strong>of</strong> the<br />
phonemes /f/, /x/ and (possibly) /r/, trigger<strong>in</strong>g <strong>of</strong> a greater importance <strong>of</strong> voic<strong>in</strong>g <strong>in</strong> the<br />
<strong>fortis</strong>/<strong>lenis</strong> opposition <strong>of</strong> obstruents (cf. Smith Stark 2000), changes <strong>in</strong> the distrubution <strong>of</strong><br />
/p/, (possibly) the creation <strong>of</strong> a greater contrast between /r/ and /l/, and phonemicization,<br />
<strong>in</strong> some dialects, <strong>of</strong> a <strong>fortis</strong> /r/.<br />
27
NOTES<br />
1 In addition to the dialects whose consonantal <strong>in</strong>ventories are presented <strong>in</strong> Table<br />
1, the phonological systems <strong>of</strong> Yalálag, Choapan, Atepec, Quioquitani, Elotepec, Zaniza,<br />
and Córdova’s (= 16 th -century Valley) <strong>Zapotec</strong> have also been exam<strong>in</strong>ed.<br />
2 I give here the consonantal <strong>in</strong>ventory <strong>of</strong> the more conservative Toledan dialect as<br />
it presents the greatest number <strong>of</strong> phonemic contrasts. While this is not essential with<br />
respect to such features as sibilant voic<strong>in</strong>g and the preservation <strong>of</strong> // as a separate<br />
phoneme (both lost by that time <strong>in</strong> the Old Castilian dialect), it is important for the<br />
contrastive status <strong>of</strong> /s (z)/ and / ()/ (which by that time had already merged <strong>in</strong><br />
Andalusian). And, even though this <strong>in</strong>ventory assumes that the old affricates /ts/ and /dz/<br />
had already given way to the correspond<strong>in</strong>g fricatives, at least two early loans provide<br />
support for their affricated pronunciation (see below). <strong>The</strong> only evidence for early yeísmo<br />
is provided by the treatment <strong>of</strong> <strong>in</strong> the word caballo ‘horse’. (On the <strong>phonology</strong> <strong>of</strong><br />
sixteenth-century American Spanish see, e.g., Lapesa 1980:282ff, Rivarola 1991:450ff,<br />
Parodi 1995:39ff, and the extensive bibliography cited there<strong>in</strong>.)<br />
3 Córdova (1578a:73) notes what amounts to dialectal variation <strong>in</strong> this respect: “A<br />
la. r. hazen que sirua de. t. vt torobaya, pro totobaya. Ciroo, pro citao” (‘they make r<br />
serve as t, as <strong>in</strong> torobaya for totobaya, ciroo for citao’).<br />
4 Spanish loans <strong>in</strong> <strong>Zapotec</strong>, unlike the contemporary borrow<strong>in</strong>gs <strong>in</strong> Nahuatl (cf.<br />
González Cazanova 1977:144, 149), Mayan (cf. Parodi 1987:346-47 and 1995:50-51)<br />
and some other Mesoamerican languages (cf. Campbell 1991:171-72; Canfield 1934:210-<br />
16) do not give evidence <strong>of</strong> a phonemic difference <strong>in</strong> the pronunciation <strong>of</strong> /b/ (< Lat<strong>in</strong><br />
-p-, b-) and // (< Lat<strong>in</strong> -v-, -b-). <strong>The</strong> outcomes <strong>of</strong> <strong>in</strong>tervocalic -b- <strong>in</strong> the word caballo <strong>in</strong><br />
28
the dialects which drop the pretonic vowel, such as Z (kwey) cannot be taken as an<br />
evidence <strong>of</strong> its fricativized pronunciation s<strong>in</strong>ce the native cluster /kb/ has the same<br />
outcome (cf. Z kwez, the Potential form <strong>of</strong> bez ‘to cry, to shout’).<br />
5 <strong>The</strong> agreement <strong>of</strong> several <strong>Zapotec</strong> dialects makes it unnecessary to analyze the<br />
second part <strong>of</strong> the word as Ca yìσ& ‘grass’ (Nellis and Hollenbach 1980:104). Borrow<strong>in</strong>g<br />
through Nahuatl or another medium is likely <strong>in</strong> this case (cf. Nahuatl colex < Spanish<br />
coles cited <strong>in</strong> González Casanova 1977:151). Comparable forms <strong>in</strong> other Mesoamerican<br />
languages are quoted <strong>in</strong> Campbell (1991:176).<br />
6 In the treatment <strong>of</strong> Spanish <strong>in</strong>tervocalic /k/ or clusters conta<strong>in</strong><strong>in</strong>g it Zaniza<br />
<strong>Zapotec</strong> dist<strong>in</strong>guishes three layers <strong>of</strong> loanwords. In the earliest layer, -/k/- > -g-, as <strong>in</strong><br />
vaca ‘cow’ > bàg. In a later layer, Sp. -/k/- > -j- (phonetically [h]), e.g. loco ‘mad’ > loj,<br />
Lucas > Lyúj, Francisco > Sijw (the treatment <strong>of</strong> Spanish /s/ here is also different from<br />
that <strong>of</strong> the earliest loans). In the most recent borrow<strong>in</strong>gs, Spanish -/k/- > Z -k-.<br />
7 In more recent loans, Spanish ch is borrowed as such <strong>in</strong> both Cajonos and Atepec<br />
(cf. Nellis and Hollenbach 104).<br />
8 <strong>Zapotec</strong> borrow<strong>in</strong>gs <strong>in</strong> general are not a good source <strong>of</strong> <strong>in</strong>formation on the<br />
voic<strong>in</strong>g <strong>of</strong> sibilants.<br />
9 Numerous other words, mostly names <strong>of</strong> objects new to Mesoamerican economy,<br />
conta<strong>in</strong> the word ‘Castilla’ as a component part, for example YB yetxtil ‘bread’ (yet<br />
‘tortilla’), za’axtil ‘pomegranate’ (za’a ‘ear <strong>of</strong> corn’), bi’oxtil ‘maggot’ (bi’o ‘flea’),<br />
yaxtil ‘tall reed’ (ya ‘reed’); ZooZ zaxtil ‘pomegranate’, yetextil ‘bread’; Co yokssttill<br />
‘soap’; A ettaxtila ‘bread’, daaxtila ‘(broad) bean’ (daa ‘(kidney) bean’), yua’xtila<br />
‘wheat’ (yua’ ‘maize’); SLQZ gueht x:tiilly ‘pan’, bihx:tiilly ‘soap’; MZ yätxtiil ‘bread’,<br />
29
säxtiil ~ zäxtiil ‘pomegranate’ (zä’ ‘ear <strong>of</strong> corn’), biäxtiil ‘soap’, bedzxtiil ‘m<strong>in</strong>t’ (bedz-<br />
‘seed <strong>of</strong> fruit’), manxanilxtiil ‘camomile sp.’.<br />
10 I thank Dr. Kev<strong>in</strong> Terraciano for suggest<strong>in</strong>g to me this etymology. Phonetically,<br />
the reduction <strong>of</strong> a four-syllable word to one syllable is not unusual (cf. T dì from Spanish<br />
melodía). As one <strong>of</strong> the strategies for adjust<strong>in</strong>g Spanish words to the mostly bisyllabic<br />
structure <strong>of</strong> <strong>Zapotec</strong> vocabulary, there is a tendency, especially <strong>in</strong> early loans, to drop<br />
pretonic syllables (e.g. Z mìzh < camisa ‘shirt’; Co wai < caballo ‘horse’; Zoo lmet, YB<br />
lmet < limeta ‘bottle’; YB/Zoo migw, Z miw < amigo ‘friend’; A yèrù < agujero ‘hole’).<br />
In some cases, only the pretonic vowel drops (MZ mbaal, SLQZ mbaaly < compadre<br />
‘godfather’; YB/Zoo xcan, MZ xcaan < escaño ‘bench with a back’; Z kwey, T káy, Q<br />
gay < caballo ‘horse’; YB/Zoo dmigw < dom<strong>in</strong>go ‘Sunday’; YB rmech, SLQZ<br />
[…]rmuudy < remedio). This tendency, however, is counterbalanced by the many<br />
<strong>in</strong>stances <strong>in</strong> which the pretonic syllable or syllables have been preserved.<br />
11 Other examples <strong>of</strong> this borrow<strong>in</strong>g chronology <strong>in</strong>clude Co (kos)ak < ajo ‘garlic’<br />
(early shape <strong>of</strong> the word preserved e.g. <strong>in</strong> Z àzh); A mécú < bermejo ‘vermilion’, necu <<br />
conejo ‘rabbit’ (the more recent loan <strong>in</strong> A is cuneju; the orig<strong>in</strong>al palatal fricative is<br />
reflected, with metathesis, <strong>in</strong> Ca znékw), yèrù < *geru < agujero ‘hole’ (g- > y-/__i, e <strong>in</strong><br />
A; a sixteenth-century treatment <strong>of</strong> the same fricative can be seen <strong>in</strong> A gúψ)á < aguja<br />
‘needle’); T ãnk, SLQZ a’nngl < angel (cf. more recent T ãhy < Ángela, SLQZ<br />
Anjalye’nn < Angel<strong>in</strong>a).<br />
12 Explanation <strong>of</strong> the treatment <strong>of</strong> Spanish obstruents <strong>in</strong> recent loans <strong>in</strong> terms <strong>of</strong><br />
coexistent phonemic systems (cf. Fries and Pike 1949) is also possible. Such an<br />
30
explanation would assume that <strong>in</strong> the speech <strong>of</strong> bil<strong>in</strong>guals Spanish obstruents are<br />
contrasted as voiced and voiceless and native obstruents as <strong>fortis</strong> and <strong>lenis</strong>.<br />
31
APPENDIX<br />
(1) Spanish loanwords were exam<strong>in</strong>ed <strong>in</strong> the follow<strong>in</strong>g <strong>Zapotec</strong> dialects:<br />
(a) Northern: Atepec (A), Cajonos (Ca), Yatzachi El Bajo (YB), Zoogocho (Zoo);<br />
(b) Central: Isthmus <strong>Zapotec</strong> (I), Mitla (MZ), San Lucas Quiav<strong>in</strong>í <strong>Zapotec</strong> (SLQZ);<br />
(c) Southern: Santa María Coatlán <strong>Zapotec</strong> (Co), Quiegolani (Q);<br />
(d) Westerm: San Lorenzo Texmelucan <strong>Zapotec</strong> (T), Zaniza <strong>Zapotec</strong> (Z).<br />
<strong>The</strong>re have been only two studies devoted to Spanish loanwords <strong>in</strong> <strong>Zapotec</strong>,<br />
Fernández’ (1965) paper on Mitla and Pickett’s (1992) article on Isthmus <strong>Zapotec</strong>. <strong>The</strong><br />
other loans on which the present study is based were culled from various dictionaries and<br />
wordlists, as follows:<br />
Atepec: Neil Nellis and Jane Goodner de Nellis. 1983. Diccionario zapoteco de Juarez.<br />
México, D.F.: Instituto L<strong>in</strong>güístico de Verano;<br />
Cajonos: Nellis and Hollenbach 1980;<br />
Yatzachi El Bajo: Butler, Inez M. 1997. Diccionario zapoteco de Yatzachi. Tucson, AZ:<br />
Instituto L<strong>in</strong>güístico de Verano;<br />
Zoogocho: Long, Rebecca C., and S<strong>of</strong>ronio Cruz M. 1999. Diccionario zapoteco de San<br />
Bartolomé Zoogocho, Oaxaca. Instituto L<strong>in</strong>güístico de Verano.<br />
Isthmus: Pickett, Velma B. 1992. Palabras de préstamo en zapoteco del Istmo. Scripta<br />
philologica <strong>in</strong> honorem Juan M. Lope Blanch, ed. Elizabeth Luna Traill, pp. 69-<br />
76. México: UNAM // Pickett, Velma et al. 1979. Vocabulario zapoteco del<br />
Istmo. México, D.F.: Instituto L<strong>in</strong>güístico de Verano;<br />
Mitla: Fernández de Miranda, María Teresa. 1965. Préstamos españoles en el zapoteco de<br />
Mitla. AINAH 17:259-73 // Stubblefield, Morris, and Carol Miller de<br />
32
Stubblefield. Diccionario zapoteco de Mitla, Oaxaca. México, D.F.: Instituto<br />
L<strong>in</strong>güístico de Verano;<br />
SLQZ: Munro and Lopez 1999;<br />
Coatlán: Rob<strong>in</strong>son 1963;<br />
Quiegolani: Regnier 1993;<br />
Texmelucan: Speck, Charles H. 1978. <strong>The</strong> <strong>phonology</strong> <strong>of</strong> Texmelucan <strong>Zapotec</strong> verb<br />
irregularity. M.A. thesis, University <strong>of</strong> North Dakota;<br />
Zaniza: author’s field notes.<br />
(2) <strong>The</strong> most recurr<strong>in</strong>g <strong>of</strong> the early Spanish loans belong to several lexical<br />
categories. <strong>The</strong> loanwords that may have entered <strong>Zapotec</strong> later than the sixteenth century<br />
are given <strong>in</strong> brackets.<br />
Plants and foodstuffs: aceite, ajo, arveja, arroz, azúcar, col, (cuajo), culantro, chocolate,<br />
(durazno), garbanzo, granada, mezcal, mole, mostaza, naranja, pan,<br />
panela, papaya, queso, sandía, trigo.<br />
Animals: caballo, cochi, conejo, chivo, macho, mico, micho, mula, nigua,<br />
novillo, pollo, (toro), vaca, yegua.<br />
Utensils and the like: aguja, almohada, batea, (bolsa), botón, canoa, clavija, cuchara,<br />
cuchillo, escaño, esqu<strong>in</strong>a, estaca, (gancho), garrote, (horno), jabón,<br />
jarro, jícara, lazo, limeta, machete, mazo, mecha, mesa, mol<strong>in</strong>illo,<br />
(muñeca), navaja, paño, plato, servilleta, silla, silla de montar,<br />
(surco), taza, teja, tijeras, timón, (tienda, t<strong>in</strong>ta, tiro, trompo).<br />
Dress: camisa, (c<strong>in</strong>cho, c<strong>in</strong>ta), media, sombrero, zapato.<br />
33
Pr<strong>of</strong>essions/<strong>in</strong>stitutions:albañil, alcalde, banco, capitán, Castilla, (ciudad), coc<strong>in</strong>ero,<br />
escuela, mozo, pastor, soldado, testigo.<br />
Money: d<strong>in</strong>ero, maravedí, medio, peso, tomín, real.<br />
Religion: anima, amigo, comadre, compadre, cruz, dios, gigante, misa,<br />
remedio, sacristán, santo, (vicio), vigilia.<br />
Calendar: dom<strong>in</strong>go, (lunes, sábado), semana, (tiempo).<br />
Adjectives and misc.: (azul), barato, hasta, (loco), manso, morado.<br />
Proper names: Andrés, José, Juan, Juana, Lucas, Luis, Luisa, mixe, Pablo, Pedro,<br />
Tomás.<br />
34
REFERENCES<br />
Avel<strong>in</strong>o, Heriberto. 2001. <strong>The</strong> phonetic correlates <strong>of</strong> <strong>fortis</strong>-<strong>lenis</strong> <strong>in</strong> Yalálag <strong>Zapotec</strong><br />
consonants. M.A. thesis, UCLA.<br />
Benton, Joseph. 1988. Proto-<strong>Zapotec</strong> <strong>phonology</strong>. Ms.<br />
Broadwell, Aaron. 2000. Fortis/<strong>lenis</strong> dist<strong>in</strong>ctions <strong>in</strong> early <strong>Zapotec</strong>an manuscripts. Paper<br />
read at the conference La voz <strong>in</strong>dígena de Oaxaca, Los Angeles, 19-20 May 2000.<br />
Campbell, Lyle. 1991. Los hispanismos y la historia fonética del español en América. El<br />
español en América: actas del III Congreso Internacionald del Español de<br />
América, ed. C. Hernández et al., vol. 1, pp. 171-179. Valladolid: Junta de<br />
Castilla y León.<br />
Canfield, Delos L<strong>in</strong>coln. 1934. Spanish literature <strong>in</strong> Mexican languages as a source for<br />
the study <strong>of</strong> Spanish pronunciation. New York: Instituto de las Españas en los<br />
Estados Unidos.<br />
Córdova, Juan de. 1886 [1578a]. Arte del idioma zapoteco. Facsimile edition. México:<br />
Ediciones Toledo.<br />
_____. 1886 [1578b]. Vocabvlario en lengva çapoteca. Facsimile edition. México:<br />
Ediciones Toledo.<br />
Diccionario zapoteco-español: reglas para el entendimiento de las variantes dialectales de<br />
la sierra hecho por los zapotecos de la variante del sector xhon. 1995.Yojovi,<br />
Solaga, Oaxaca: Zanhe Xbab SA, A.C.<br />
Fries, Charles C., and Kenneth L. Pike. 1949. Coexistent phonemic systems. Language<br />
25:29-50.<br />
González Casanova, Pablo. 1977. Estudios de l<strong>in</strong>güística y filología nahuas, ed.<br />
Ascensión H. de León-Portilla. México: UNAM.<br />
Jaeger, Jeri J. 1983. <strong>The</strong> <strong>fortis</strong>/<strong>lenis</strong> question: evidence from <strong>Zapotec</strong> and Jawoñ. Journal<br />
<strong>of</strong> Phonetics 11:177-89.<br />
Kaufman, Terrence. 1994. Proto-<strong>Zapotec</strong> reconstructions. Ms.<br />
_____. 1983. New perspectives on comparative Otomanguean <strong>phonology</strong>. Ms.<br />
_____. n.d. <strong>The</strong> <strong>phonology</strong> and morphology <strong>of</strong> <strong>Zapotec</strong> verbs. Ms.<br />
Lapesa, Rafael. 1980. Historia de la lengua española. 8 th ed. Madrid: Gredos.<br />
35
Manrique Castañeda, Leonardo. 1966-67. El zapoteco de fray Juan de Córdoba. Anuario<br />
de Letras 6:203-11.<br />
Marlett, Stephen A., and Velma B. Pickett. 1987. <strong>The</strong> syllable structure and aspect<br />
morphology <strong>of</strong> Isthmus <strong>Zapotec</strong>. IJAL 53:398-422.<br />
Munro, Pamela, and Felipe H. Lopez. 1999. Di’csyonaary X:tèe’n Dìi’zh Sah Sann<br />
Lu’uc/San Lucas Quiav<strong>in</strong>í <strong>Zapotec</strong> dictionary/Diccionario zapoteco de San Lucas<br />
Quiav<strong>in</strong>í. UCLA Chicano Studies Research Center.<br />
Nellis, Donald G., and Barbara E. Hollenbach. 1980. Fortis versus <strong>lenis</strong> <strong>in</strong> Cajonos<br />
<strong>Zapotec</strong> <strong>phonology</strong>. IJAL 46:92-105.<br />
Parodi, Claudia. 1995. Orígenes del español americano. Vol. 1. Reconstrucción de la<br />
Pronunciación. México: UNAM.<br />
_____. 1987. Los hispanismos en las lenguas mayances. Studia humanitatis: homenaje a<br />
Rubén Bonifaz Nuño, ed. A. Ocampo, pp. 339-49. México: UNAM.<br />
Pickett, Velma B., et al. 1998. Gramática popular del zapoteco del Istmo. Juchitán,<br />
México: Centro de Investigación y Desarrollo B<strong>in</strong>nizá A.C. and Tucson, AZ:<br />
Instituto L<strong>in</strong>güístico de Verano.<br />
Regnier, Sue. 1993. Quiegolani <strong>Zapotec</strong> <strong>phonology</strong>. SIL-University <strong>of</strong> North Dakota<br />
Work Papers 37:37-63.<br />
Rivarola, José Luis. 1991. En torno a los orígenes del español de América. Scripta<br />
philologica <strong>in</strong> honorem Juan M. Lope Blanch, ed. Elizabeth Luna Traill, vol. 1,<br />
pp. 445-68. México: UNAM.<br />
Rob<strong>in</strong>son, Dow F. 1963. Field notes on Coatlan <strong>Zapotec</strong>. Hartford Sem<strong>in</strong>ary Foundation.<br />
Smith Stark, Thomas C. 2000. La ortografía del zapoteco en el Vocabvlario de fray Juan<br />
de Córdova. Ms.<br />
Suárez, Jorge A. 1973. On Proto-<strong>Zapotec</strong> <strong>phonology</strong>. IJAL 39:236-49.<br />
Swadesh, Morris. 1947. <strong>The</strong> phonemic structure <strong>of</strong> Proto-<strong>Zapotec</strong>. IJAL 13:220-30.<br />
36